Title: Identification of relevant features influencing movie reviews using sentiment analysis

Authors: Isha Gupta; Indranath Chatterjee; Neha Gupta

Addresses: Faculty of Computer Applications, Manav Rachna International Institute of Research and Studies, Faridabad – 121003, India ' Department of Computer Engineering, Tongmyong University, Busan – 48520, South Korea ' Faculty of Computer Applications, Manav Rachna International Institute of Research and Studies, Faridabad – 121003, India

Abstract: Sentiment analysis is a systematic text mining research that examines individuals' behaviour, approach, and viewpoint. This paper analyses viewers' sentiments towards the movies released during the pandemic. This study employs the sentiment analysis techniques on movie reviews' accessed in real-time from internet movie database (IMDb). The paper's main objective is to identify the potential words that contribute to the biases of the reviews and influence overall viewers. The proposed methodology has employed valence aware dictionary for sentiment reasoning based on sentiment analysis of overall reviews, followed by application to various movie genres. Finally, we have applied Pearson's correlation analysis to find the association between the words among the genres. The paper also calculates the sentiment scores of reviews using different sentiment analysis models. Our results showed a minimum of 17% features common genre-wise. It reveals sets of most distinct influential words, which may be vital for understanding the nature of the language used for a particular kind of movie.

Keywords: sentiment analysis; feature selection; sentiment scores; internet movie database; IMDb reviews; adjectives and adverbs features.

DOI: 10.1504/IJDMMM.2023.131395

International Journal of Data Mining, Modelling and Management, 2023 Vol.15 No.2, pp.169 - 183

Received: 05 Jan 2022
Received in revised form: 26 Jun 2022
Accepted: 27 Jun 2022
Published online: 09 Jun 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article